Xilinx Automotive Optical Flow Solution Addresses Image Processing Requirements of Advanced Driver Assistance Systems
At the Electronica 2008 Conference, Xilinx debuted its XilinxAutomotive (XA) Optical Flow solution for vision-based driver assistance (DA) systems.
XA FPGAs with DDC high performance image processing IP demonstrates a cost-effective programmable development platform for camera-based DA applications
At the Electronica 2008 Conference, Xilinx debuted its XilinxAutomotive (XA) Optical Flow solution for vision-based driver assistance (DA) systems. This field programmable gate array (FPGA) implementation integrates high- performance image processing capabilities from Digital Design Corporation (DDC) into a cost-effective, programmable platform that can be customized to support a variety of advanced applications, including pedestrian detection, collision warning, sign recognition, blind spot detection, back-up and parking aid.
The XA Optical Flow solution delivers the scalable processing horsepower needed to implement pixel-level image processing functions that extract relative object motion information from a sequence of video frames. Its FPGA implementation provides a number of adjustable parameters to support performance tuning tailored to the requirements of the target application.
The emerging automotive market for camera-based driver assistance systems is expected to grow from $941 million to $2587 million starting in 2010 to 2014. Helping to fuel this growth is the automotive industry's evolution from simple camera systems that provide images to the driver to more sophisticated collision warning systems that can notify the driver of potential crash situations.
Superior parallel processing performance coupled with cost-effective functional density make XA FPGAs an attractive programmable platform for automotive system developers. Xilinx’s new XA Optical Flow solution goes even further, leveraging DDC's image processing technologies as key building blocks for FPGA-based DA systems. Developers are no longer constrained by the limitations of serial DSPs which can run out of steam for high-end pixel processing, or ASIC and ASSP alternatives that don't have the flexibility to support rapid integration of new algorithms and application requirements as the market evolves.
XA Optical Flow Solution
The XA Optical Flow solution is built on a Xilinx Spartan(R)-3 device and DDC Optical Flow IP core with its innovative motion estimation algorithms. The Optical Flow IP core extracts relevant information about the roadway environment from the video images provided by vehicle-based camera systems. The extracted motion images are segmented into separate objects (pixel blocks) and processed using a block matching technique. The output is a two- dimensional vector map that can be utilized to assess collision threats.
The block matching function requires a very high number of operations per second to perform sum of absolute difference (SAD) calculations across image frames. The Xilinx FPGA executes these operations in parallel, so in a single clock cycle multiple operations can be carried out. This is a higher performance, more efficient optical flow solution than is possible with traditional digital signal processing (DSP) CPUs, which generally operate serially and require very high clock rates.
As the image resolution of vehicle-based cameras increases, the amount of data generated and throughput required to process it also increases greatly. This imposes a heavy burden on existing DSP CPU-based approaches, which cannot keep up with performance requirements. The combination of Xilinx FPGAs and high-end optical flow core provides a scaleable processing solution with the speed needed for sophisticated vision-based driver assistance applications.